PREDICTING AND OPTIMIZING OF SURFACE ROUGHNESS IN METAL CUTTING PARAMETERS

Abstract

This study aimed to predict and optimize the surface roughness for work piece type (St 42crmo4) in straight turning process, the study will focus on three cutting parameters that effect on the surface roughness which is the cutting speed feed rate and depth of cut while maintaining the other parameter constant, to predict and optimize the response two models are developed, the first is mathematical second order model by using response surface methodology to analyze the cutting parameter effects on surface roughness , and the second is artificial Neural Networks model to predict and optimize the response, the experiments were conducted by three level full factorial design methodology in (CNC) lathe machine type (TB-15Z ~ NL635SCZ), the response variable namely the surface roughness was measured using Portable surface roughness tester (Surf-test SJ-210 SERIES), the effect of process parameters with the output variable were predicted which indicates that the cutting speed has significant role in producing least surface roughness followed by feed and at least depth of cut, and the optimized parameters which give the optimal response value are gained by utilizing (ANN) model.